Polymer Durability and Radiation Effects - American Chemical Society

---

Chapter 5

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

Mechanism of Action of Hindered Amines as Long-Term Heat Stabilizers Pieter Gijsman

1,2

and Mikael Hamskog

2

1

DSM Research, P.O. Box 18, NL-6160 MD Geleen, The Netherlands University of Kristianstad, Kristianstad, Sweden 2

Although Hindered Amine Stabilizers (HAS) originally only were used as UV-Stabilizer, it is nowadays well known that they act as stabilizers against the thermo-oxidative degradation of polypropylene (PP) too. Their mechanism of action is different from the traditionally used phenolic antioxidants (PAO). A comparison was made between the effectivity of a PAO and a HAS as long-term heat stabilizer. It is found that both types of stabilizer interfere in different ways in the degradation of PP. These differences will be clarified using results on the degradation of unstabilized PP and model experiments with decalin, decalin / lauryladehyde mixtures and squalane. For oligomeric HAS types containing different aliphatic amines in their backbone, differences in activity were found. For these HAS types, it is shown that parts of the backbone can act as stabilizer too.

© 2008 American Chemical Society

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

49

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

Introduction Although Hindered Amine Stabilizers (HASs) originally only were used as UV-Stabilizer, it is nowadays well known that they act as stabilizers against the thermo-oxidative degradation of polypropylene (PP) too. However, their mechanism of action is different from the traditionally used phenolic antioxidants (PAOs). It was shown that polymers containing HASs and PAOs show a difference in decline of the mechanical properties during ageing. For PP containing phenolic antioxidants a sudden decline after a period of almost no changes is found, while PP containing HASs show a more or less linear decline intime [1,2,3]. A comparison between a HAS and a PAO as stabilizers against the oxidation of PP showed that they behave differently [4]. Based on experiment showing the ability of the HAS and the PAO to reduce the oxidation rate of decalin, squalane and decalin/lauryl aldehyde mixtures, it is shown that the HAS is mainly active in the presence of aldehydes. For unstabilized PP it is shown that aldehydes play an important role in its oxidation [5]. These results are used to propose an oxidation mechanism for PP and a mechanism underlying the action of HAS. It was also shown that the effectiveness of HAS is not only related to their piperidinyl group. For oligomeric HAS types with deviations in chemical structure in the backbone, differences in activity were found. These differences were explained by studying the influence of parts of the backbone on the degradation mechanism [6].

Thermo-oxidation of Stabilized PP Films [4] Stabilized PP films (thickness: 150μιη) containing a high molecular weight HAS or a PAO (for chemical structure see Figure 1) were aged at 120°C in a Heraeus D-6450 Hanau oven. In all cases 0.1 wt.% Butylated Hydroxy Toluene was added to prevent degradation during processing. For more details about the polymer and the preparation of the films see ref. 4. Changes in the visual appearance of films were recorded by scanning ovenaged films. The images were made on a black background. During degradation parts of the films became white (see Figures 2 and 3), these parts are brittle. The behavior of the samples depends on the type of stabilizer. In the phenolicantioxidant-containing sample the degradation starts after 560 hours at the edges and propagates rapidly into the film (see Figure 2). After 682 hours more than 50% of the film is totally brittle. The HAS-containing film behaves differently. Degradation starts more randomly and does not spread in the way observed in the PAO-containing

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

HAS-2 (Chimassorb® 119)

Figure 1. Stabilizer chemical formulae and trade names.

HAS-1 (Chimassorb®944)

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

51

Figure 2. Images, scanned against a black background, of 150 jum thick PP films (4x4 cm2) containing 0.1% PAO-1 after oven ageingfor different times at 120°C.

samples (see Figure 3). In the presence of HAS-1 parts of the film became white after only 300 hours, but the whitening spread very slowly. Even after 1100 hours the film was still not completely white. The nature of the whitening also differed: in PP stabilized with HAS-1 the white parts were less brittle than in PP stabilized with PAO-1.

Figure 3. Images, scanned against a black background, of 150 jam thick PP films (4x4 cm2) containing 0.1% HAS-1 after oven ageing for different times at 120°C.

Imaging Luminescence (CL) of Preoxidized Stabilized PP CL intensity-time curves were measured at 135°C for at the same temperature preaged PP samples. The plaques containing PAO-1 were preaged for 640 hours (for more details see ref. 7) and the compound containing HAS-1 for 2176 hours. For the PAO containing PP, chemiluminescence is only observed after an induction period. In Figure 4 the induction times for samples taken out along a horizontal line in relation to the distance from the left edge of the plaque is plotted. From this plot it is clear that in this case the degradation is very heterogeneous (for details see ref. 7).

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

52

Figure 4. OIT values (measured at 135°C) from three experiments, each with 11 PP samples taken out along a horizontal line (see inserted picture). The PP samples were with PAO-1 stabilized and preagedfor 640 h at 135°C.

The CL intensity-time curves for the HAS containing PP looks totally different. In this case a continuous low luminescence level is observed without an induction period, which is homogenous over the whole plaque (Figure 5).

Oxidation of Decalin, Squalane and a Mixture of Lauryl aldehyde and Decalin [4] The influence of the stabilizers mentioned in Figure 1 was studied by oxidizing three different model compounds: decalin (a cyclic hydrocarbon which, like PP, contains secondary and tertiary carbons), squalane (a high molecular weight aliphatic hydrocarbon which also contains secondary and tertiary carbons), a mixture of 10% lauryl aldehyde in decalin, which could serve as a model for partly oxidized PP. All reactions were initiated with 1% of t-butylhydroperoxide and performed at 120°C (for details see ref. 4). HAS-1 has almost no effect on the oxidation of decalin (Figure 6), there is no induction period, and only a small reduction of the oxidation rate was found. The effect of PAO-1 is much greater. In the presence of this stabilizer the oxidation of decalin shows an induction time (Figure 6). The oxidation rate after the induction period is comparable to that of the control, probably because the stabilizer is consumed during the induction period. The addition of lauryl aldehyde (10%) to decalin has a tremendous effect on the oxidation rate as well as on the effectivity of the stabilizers. This mixture already starts to oxidize while it is being heated to its final oxidation temperature (120°C), which is the reason that the data are not shown in Figure 7. In this case the effectivity of PAO-1 and HAS-1 is comparable.

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

53

Figure 5. Luminescence intensity (at 135°C) for 11 PP samples containing HAS-1 taken out along a horizontal line (in insert distance from the left edge is shown). The PP was preagedfor 2176 h at 135°C. 400

1000 Time (min)

Figure 6. Influence of 0.1 wt. % stabilizer on the oxidation rate ofdecalin at 120°C.Control (+), HAS-1 (A) andPAO-1 (o). Bars show the standard deviation.

The oxidation of squalane is slower than the oxidation of the mixture of decalin and lauryl aldehyde (compare Figures 7 and 8). Although squalane is a hydrocarbon similar to decalin, in squalane HAS are capable of reducing the oxidation rate effectively, while they were not effective against the oxidation of decalin. In squalane HAS-1 even outperformed PAO-1 (Figure 8).

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

54

ο Ε Ε ϊ (Ό α. Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

3 C 03 CD

0

5

10

15

20

25

30

35

40

Time (min)

Figure 7. Influence of 0.1 wt. % stabilizer on the oxidation rate of a decalin/lauryl aldehyde mixture at 120°C HAS-1 (Èl) and PAO-1 (o). Bars show the standard deviation.

0

2000

4000

6000

8000

Time (min)

Figure 8. Influence of 0.1 wt. % stabilizer on the oxidation rate ofsqualane at 120°C. Control (+), HAS-1 (L) and PAO-1 (o). Bars show the standard deviation.

It is clear that the efficiency of the different stabilizers highly depends on the model system. PAOs are effective in all systems, but HAS are not effective in decalin. The oxidation of decalin mainly takes place at the tertiary carbon, leading to cyclodecanone and cyclodecanol [8]. This oxidation does not lead to the formation of aldehydes and acids. When aldehydes are added to decalin HAS becomes effective, so HAS is effective against the oxidation of aldehydes. HAS is also effective against the oxidation of squalane, which, like decalin, is a hydrocarbon containing secondary and tertiary carbons. However, because of its

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

55

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

non-cyclic structure the oxidation of squalane can lead to aldehydes, thus HAS is effective when aldehydes are added or can be formed. The above-mentioned results are all valid for model experiments in the liquid phase. The question still is: what is their relevance for the mechanism underlying the long-term heat-stabilizing action of HAS in PP? After all, the above-mentioned positive effects of HAS can only be of relevance when aldehydes play a major role in the oxidation mechanism of PP. In the following an attempt is made to show the important role played by aldehydes in the low temperature oxidation of unstabilized PP.

Oxidation of Unstabilized PP The long-term thermal degradation of PP is due to an autoxidative process [9,10]. In this mechanism the decomposition reactions of the formed peroxides play an essential role. A possible way of studying this reaction is to determine the decrease in the concentration of peroxides of oxidized PP in an inert medium as a function of time (see Figure 9). It is shown that this decomposition takes place in two stages. Initially a fast decomposition was observed, which was followed by a slow decomposition. This type of decay could not be described by first or second order reaction. Assuming that the titrated concentration of peroxides consists of twofractions(both decomposing according to first order), the calculated data could be fitted with the experimental values (see Figure 9). The calculated k values are 4.2* 10" per second for the fast decomposition reaction and 1.8*10" per second for the slow decomposition reaction. Besides the k values, the amounts of rapidly decomposing and slowly decomposing peroxides at the beginning of the decomposition were calculated using this model. As the peroxides are introduced into the polymer by first oxidizing the polymer, the concentrations of peroxides as a function of oxidation time can be calculated. From these calculations it can be concluded that at the beginning of the oxidation (induction period) mainly slowly decomposing peroxides are formed. The influence of both peroxides on the oxidation rate was determined in an oxidation (0 )-decomposition (N )-oxidation (0 ) experiment (Figure 10). During the oxidation the concentrations of both peroxides were built up, after several oxidation times the oxidized powder was placed in nitrogen and a decomposition experiment was started. In the last stage of this experiment, the influence of varying the ratio between the two types of peroxide on the oxidation rate was determined. Especially for the samples with the longest decomposition time a reduction in oxidation rate was found. Although these samples still contains (slow decomposing) peroxides an induction period is found (Figure 10). In literature there is general agreement that peroxides in PP consist of a rapidly decomposing fraction and a slowly decomposingfraction,but there is no agreement on the origin of these fractions [11,12,13,14] According to 6

7

2

2

2

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

56

2000

1000 Time (hrs)

Figure 9. Peroxide concentrations as calculated according to the abovementioned equations (lines) and determined by an iodometric method (points) during a peroxide decomposition experiment at 70°C with polypropylene pre-oxidizedfor several times at 70°C. 3000

3 2 αί

2000

α D I

100

c

250

500 , decomposition time (hrs)

- 300

v

CO

1 α

- 200

ê

100 1000 250

500

750 T o t a l time

1000

1250

(hrs)

Figure 10. Oxygen uptake of PP at 70°C after 380 h of oxidation and 0 ( 1), 25,(2), 50 (3), 100 (4), 250 (5) and500 (6) h decomposition in nitrogen. Also shown are the measured peroxide concentration(lS) and the calculated concentration of rapidly decomposing peroxides (+) after the above mentioned decomposition times.

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

57

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

Zahradnikova et al. dimethyl sulphide (DMS) reacts much faster with peracids than with other types of peroxides [14]. Although there still is discussion about this method, from peroxide determination before and after treatment with DMS it was concluded that the fast decomposing peroxides are peracids [15], resulting from aldehydes which are formed via the decomposition of hydroperoxides .

Mechanism of Degradation of PP and Mechanism Underlying the Action of PAOs and H A S It was shown that the radical formation rate of hydroperoxides (slowly decomposing peroxides) is too slow to explain the increase in oxidation rate (Figure 10). This increase in oxidation rate is ascribed to the decomposition of peracids that are formed as a result of the decomposition of hydroperoxides. This oxidation is so fast that it is hard to stop by antioxidants, so even in the presence of a PAO an oxidation can no longer be stopped once it has been initiated, leading to an infectious spreading of the degradation as shown in Figures 2 and 4. Model experiments with different hydrocarbons and aldehydes showed that HAS are not capable in reducing the oxidation of hydrocarbons, but they are capable in reducing the oxidation rate in the presence of aldehydes and prevent the formation of peracids. So in the presence of HAS the oxidation is not accelerated and the oxidation does not spread, although the primary oxidation is not stopped, which means that the polymer oxidizes from the beginning resulting in a homogeneous degradation and continuous low luminescence levels as shown in Figures 3 and 5.

Influence of Amines on the Oxidation of PP [6] Oligomeric HAS types with a difference in the amines present in their backbone show a difference in activity. In the following it is shown that this can be due to a stabilizing effect of the backbone by showing the stabilizing effect of different amines. The addition of an aliphatic amine did not have a major influence on the starting time of the degradation, but it had an influence on the spreading of the oxidation (Figure 11). As the spreading rate is determined by the oxidation of formed aldehydes, it is reasonable to assume that the aliphatic amines counteract the influence of these aldehydes. It is known that aliphatic amines can react with aldehydes to form more stable imines or enamines [16], which is a reasonable mechanism that explains the activity of aliphatic amines. Thus for oligomeric HAS with aliphatic amines in their backbone not only the piperidinyl groups is important, but also the chemical structure of the backbone.

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.

Downloaded by IMPERIAL COLLEGE LONDON on February 18, 2015 | http://pubs.acs.org Publication Date: December 21, 2007 | doi: 10.1021/bk-2007-0978.ch005

58

Figure 11. Scanned images of 150-jum PPfilms containing: 0.1 wt.% HAS-2 without additional additives (top), with 0.1% Distearylamine (middle) and with 0.1% of the stearic acid salt of N,N',N"-trimethylbis (hexamethylene)triamine (bottom) aged for different times at 120°C

Acknowledgements Sabic-Europe b.v. is gratefully acknowledged for allowing us to publish the results presented in this paper.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

Gugumus, F. Polym. Degrad. Stab. 1994,44,299-322. Gensler R.; Plummer, C. J. G.; Kausch H. H.; Kramer, E.; Pauquet, J. R.; Zweifel, H. Polym.Degrad. Stab. 2000,67, 195-208. Gijsman, P. Polym. Degrad. Stab. 1994,43, 171-176. Gijsman, P.; Gitton, M. Polym. Degrad Stab. 1999,66,365-371. Gijsman P.; Hennekens J.; Vincent, J. Polym. Degrad Stab. 1993,42,95-105. Gijsman, P.; Gitton, M. Polym. Degrad. Stab. 2003,81,483-489. Hamskog, M.; Terselius, B.; Gijsman, P. Polym. Degrad. Stab. 2003, 82, 181186. Gijsman, P.; Internal DSM Report, 1985. Gugumus F. in Plastic Additives Handbook, Eds. Gachter R., Muller H., Hanser Publishers, Munich, Vienna, New York, 1990, pp. 1-104. Gugumus F. in Oxidation Inhibition In Organic Materials, Vol.1, eds. Pospisil J.; Klemchuk, P. P., CRC Press, Inc. Boca Raton, Florida, 1990, pp. 61-172. Chien J. C. W.; Jabloner H. J. J. Polym. Sci. Part: A-l 1968,6,393-402. Zolotova, N. V.; Denisov, E. T. J. Polym. Sci. Part: A-l 1971, 9,3311-3320. Shlyapnikov, Yu. A.; Bogaevskaya, T. A.; Kiryushkin, S. G.; Monakhova, T. V. Eur. Pol. J. 1979,15, 737-742. Zahradnikova, A.; Sedlar J.; Dastych, D. Polym. Degrad. Stab. 1991, 32, 155176. Gijsman, P.; Kroon, M.; van Oorschot, M. Polym. Degrad. Stab. 1996, 51, 313. Morrison and Boyd, Organic Chemistry, 3rd edition, pp.858-859.

In Polymer Durability and Radiation Effects; Celina, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2007.